Translator turns actual dolphin sounds into English

In a development that would excite Doctor Doolittle, scientists have developed a working translator that can take dolphin sounds and turn them into spoken English. The translator called CHAT (Cetacean Hearing and Telemetry), takes the whistling sounds that dolphins make to communicate, and matches them to a known database of meanings.

Before you get too excited, the recently reported translation was of just one word — sargassum — a type of seaweed, so it's not as if the scientists are suddenly having long chats with the dolphins about their family lives.

CHAT was developed by Thad Starner, one of the lead developers behind Google Glass, working in tandem with Denise Herzing of the Wild Dolphin Project. The smartphone-sized computer is worn by a diver in the water with the dolphins, and is connected to LEDs inside the diver's mask that lets them know the direction of any incoming sounds. A small handheld controller called a twiddler acts as both a mouse and keyboard, allowing the diver to select appropriate sounds that can be played back in real time in response to the dolphins sounds. Herzing hopes this will eventually lead to actual two-way conversations with the dolphins.

Listening to dolphin sounds isn't as easy as dropping a microphone in the water, as their vocal range extends as high as 200 kHz, or more than ten times the maximum upper frequency of human hearing. To make it easier to hear when the dolphins are trying to communicate, the researchers have taught them to use new lower-pitched sounds for their interactions with humans. While the translated "sargassum" word followed the exact pattern they had taught the dolphins to use, it was at a much higher pitch than expected.

Where this will eventually lead is hard to determine, but anthropologist and neuroscientist Terrence Deacon from the University of California at Berkeley says that better communications are key to our further understanding of animal cultures.

Herzing and Starner will present their latest findings at the Speech and Signal Processing Conference in Florence, Italy next month.

New Scientist, via CNET

For the latest tech stories, follow DVICE on Twitter
at @dvice or find us on Facebook